© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 26585
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Expand."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    164
    +0.13
    0.7%
    286
    +0.12
    0.7%
    484
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    286
    +0.13
    0.01
    343
    +0.12
    0.01
    449
    +0.10
    0.01
    Negative Logits
    reet
    -1.70
     worry
    -1.68
     proud
    -1.66
     thrive
    -1.65
     complete
    -1.64
     great
    -1.61
     ever
    -1.60
     believe
    -1.60
     thing
    -1.58
     find
    -1.57
    POSITIVE LOGITS
    ingen
    1.86
    ibilit
    1.78
    ing
    1.68
    aggio
    1.66
    nement
    1.66
    endering
    1.64
    yscall
    1.63
    notes
    1.63
    ÃŃvel
    1.58
    evol
    1.57
    Activations Density 0.015%

    No Known Activations